This project will study complex signal processing in the cochlear nucleus where the first steps of a cascade of parallel and hierarchical auditory signal processing events take place. Distinct anatomical cell types, each with different targets, form the substrate for parallel pathways and along with a rich network of intranuclear circuits shape the outputs of projecting cells. The present aims are: to make structure/function correlations that allow physiological identification of anatomical cell types and to explore how these types process complex sounds including amplitude modulation, N-harmonic component, and speech stimuli. Extracellular-and intracellular-recording will be combined with neurobiotin labeling to establish structure- function relations for five groups of cells that are not well understood: (1) Onset cells in the PVCN: What cells group(s) are responsible for O/I and O/L patterns and what is their relation to O/C cells? (2) The primary-like responses, PL, PL/N, and O/L, in the AVCN arise from bushy cells but distinguishing between them is often difficult based on responses to tones alone. Do these patterns arise from bushy-cells exclusively? (3) Establish the morphological basis for sustained and transient choppers in the AVCN. (4) Establish whether onset-graded response patterns, units with type II receptive fields and tuberculoventral cells in the DCN are one and the same cell group? (5) Establish whether cartwheel cells in the DCN respond to tones clicks, AM, noise, noise with notches, and/or contra-lateral acoustic stimulation. A new initiative will investigate neurophysiological responses to a diverse set of speech sounds including sounds that have not been used in previous neurophysiological studies. Examining a large array of speed sounds will make possible the extrapolation of results to speech perception. Psychophysical capabilities will be studied in the chinchilla as a behavioral adjunct to investigations of neural responses to speech using the same speech sounds that will be presented in the neurophysiological studies. In many cases, neural responses to stimulus pairs, drawn from series of speech stimuli varying from one phoneme to another, that bracket the animal's behaviorally derived thresholds will be most intensely studied. Information-theoretic computational analyses of neural responses will be used to enable direct inference from neural encoding to sensory capacity.